adaptive-communication
Use when detecting ambiguous user intent, hedging language, open-ended framing, personal context before requests, or when unsure whether user wants exploration vs direct answer. Applies to all conversations.
43
42%
Does it follow best practices?
Impact
—
No eval scenarios have been run
Passed
No known issues
Optimize this skill with Tessl
npx tessl skill review --optimize ./adaptive-communication/skills/adaptive-communication/SKILL.mdQuality
Discovery
14%Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.
This description focuses entirely on trigger conditions (the 'when') without ever explaining what the skill actually does (the 'what'). The phrase 'Applies to all conversations' makes it extremely broad and conflict-prone, and the trigger terms describe internal cognitive states rather than natural user language. The description needs a fundamental rewrite to specify concrete actions and narrow its scope.
Suggestions
Add explicit capability statements describing what the skill does, e.g., 'Asks clarifying questions to disambiguate user intent' or 'Provides structured exploration of multiple interpretations before answering.'
Remove or significantly narrow 'Applies to all conversations'—specify the particular domains or conversation types where this skill adds value over default behavior.
Reframe trigger terms around observable user behaviors or phrases rather than abstract concepts, e.g., 'Use when user says things like "maybe," "I'm not sure," "what do you think about," or asks broad questions without clear success criteria.'
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | The description lacks concrete actions. It mentions 'detecting ambiguous user intent' and 'hedging language' but never states what the skill actually does once triggered. There are no specific capabilities listed—no verbs describing outputs or transformations. | 1 / 3 |
Completeness | The 'when' is addressed with trigger conditions, but the 'what does this do' is entirely missing—there is no explanation of what action the skill performs or what output it produces. Additionally, 'Applies to all conversations' is overly broad and undermines the 'when' guidance. | 1 / 3 |
Trigger Term Quality | It includes some relevant trigger concepts like 'ambiguous user intent,' 'hedging language,' 'open-ended framing,' and 'exploration vs direct answer,' but these are meta-cognitive terms unlikely to appear in natural user queries. Users wouldn't say 'I have hedging language'—they'd just hedge. | 2 / 3 |
Distinctiveness Conflict Risk | 'Applies to all conversations' explicitly makes this skill maximally generic and conflicting. Any skill could overlap with something that triggers on ambiguous intent across all conversations, making it nearly impossible to distinguish from other conversational skills. | 1 / 3 |
Total | 5 / 12 Passed |
Implementation
70%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This is a well-structured behavioral skill that provides clear signal detection criteria and response strategies for adaptive communication. Its main weakness is that some guidance is somewhat obvious to Claude and could be tightened, and the actionability would benefit from concrete before/after conversation examples rather than abstract adaptation instructions. The workflow and organization are strong for this type of instruction-only skill.
Suggestions
Add 1-2 concrete before/after conversation examples showing the same ambiguous user message handled with and without adaptive communication to make the skill more actionable.
Trim anti-patterns and edge cases that Claude would naturally handle (e.g., 'don't be patronizing') to improve conciseness—focus only on non-obvious behavioral corrections.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Generally efficient with good use of tables and structured formatting, but some content is somewhat obvious to Claude (e.g., the concept of high-context vs low-context communication, the idea of acknowledging emotions before tasks). The anti-patterns section tells Claude things it likely already knows about not being patronizing. | 2 / 3 |
Actionability | Provides concrete example phrases for clarification and specific signal detection tables, but the guidance is largely behavioral/conversational rather than executable. The response adaptations are somewhat vague ('match relational tone,' 'offer scaffolding') without showing full example exchanges demonstrating the before/after of adaptive communication. | 2 / 3 |
Workflow Clarity | For a behavioral/instruction-only skill, the workflow is clear: detect signals → classify context type → apply corresponding response adaptation. The quick reference at the end provides an unambiguous decision tree. The intent clarification triggers section adds clear checkpoint logic for when to ask vs. proceed. | 3 / 3 |
Progressive Disclosure | For a standalone skill under 80 lines with no need for external references, the content is well-organized into logical sections (detection, response, edge cases, anti-patterns, quick reference) with clear headers and appropriate use of tables. No bundle files are needed for this type of behavioral guidance. | 3 / 3 |
Total | 10 / 12 Passed |
Validation
100%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 11 / 11 Passed
Validation for skill structure
No warnings or errors.
- Repository
- bencium/bencium-marketplace
- Commit
2e2fa33
- Reviewed
Table of Contents
Is this your skill?
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.